Understanding underlying trend and using it in Modeling of agricultural production function with respect to the water input

Due to the being in the desert bar and ecosystem sensitivity to water , water management in the agricultural sector of the Iran is one of the priorities of managers and planners of this section. Therefore accurate modeling of the production function, for policies to increase efficiency and reduce the use of inputs, as well as forecasts for the timely supply with low cost, is of utmost importance. In this study, the production function in the agricultural sector of Iran has been modeled using underlying trend concept, creating a state - space model and Kalman filter algorithms. We used annual time series data for the period 1974-2012. The results indicate that underlying trend is smooth and Non-linear. According to likelihood ratio statistic, the most appropriate structure for hyper parameters is random static levels with No-slope trends (Local Level Model). Capital, labor and energy elasticity were less than unity, But water elasticity was larger than unity which is obtained in the first production area. This means that with a one percent increase in water inputs causes a decrease in production of more than one percent. Because, Iran is faced with limited water resources, cannot increase the consumption of water in this sector. Thus, recommended the replacement of new irrigation systems to compensate for water deficit and also in agricultural production by needs less water. Also, the development of basic technologies (changing the agricultural system) can reduce the use of inputs and compensate the lack of water in the agricultural sector.